Everyone measures the room, almost nobody measures the return to work

The training industry has built an elaborate apparatus around the delivery of learning and a very thin one around whether any of it survives contact with a Tuesday morning. That imbalance is not accidental.

There is a moment at the end of a well run training session when the room feels different from how it felt at nine in the morning. People are talking to each other. Somebody has had an idea they want to try. The facilitator is being thanked. It is a genuinely pleasant moment, and it is also the moment at which the overwhelming majority of training evaluation takes place. A form is handed out, or a link is sent, and the participants record how they feel while they are still feeling it. The data collected at that point is real data about something, but the something is not whether the training will change any behaviour at work. It is a measure of the experience of being trained.

Three weeks later the picture is different and nobody is looking at it. The participant has returned to a workload that did not pause while they were away. The new approach they were shown requires a form that does not exist yet, or a conversation with a colleague who was not in the room, or simply more time on a first attempt than the old way takes on the hundredth. Under those conditions the old way wins, not because the person is lazy or the training was bad, but because the old way is available and the new one is not. The skill has not been lost. It has been shelved.

This is the transfer problem, and it is the single most consequential question in the field. Everything else that training people argue about, the design of activities, the length of sessions, the quality of materials, the platform, the room, matters only insofar as it affects what someone does afterwards. If nothing changes at work, the training was an expensive way of giving people a day out. That is a blunt sentence and most practitioners will flinch at it, but it is also the sentence that the better practitioners say quietly to each other at the end of a long year when the reporting is done.

The honest observation is that a great deal of training changes nothing, and that most of the people involved half know this. The buyer knows the request came from an incident rather than an analysis. The facilitator knows the group included several people who were told to attend and would rather have been anywhere else. The participants know that the systems they are going back to will not accommodate the thing they were just taught. Everybody performs the ritual anyway, because the ritual is what is funded and because performing it is easier than saying out loud that the problem is not a training problem at all.

What makes transfer worth writing about is that it is not mysterious. The conditions that make a new practice stick are reasonably well understood in general terms, they are not especially exotic, and most of them are cheap. Whether the person had any say in coming. Whether the new way is physically possible where they work. Whether their supervisor knows what they learned and expects to see it. Whether the practice they did in the session resembled the messy real task rather than a tidied version of it. Whether they used it within days rather than months. None of that requires new technology or a larger budget.

What it does require is a willingness to look at the part of the process that sits outside the training room, which is where most of the causes are and where trainers have the least authority. This publication exists to look there anyway: at the return journey, the fortnight after, the manager who never asks, the workplace that cannot absorb a change and paid for one regardless. It is not a comfortable subject, and it is not served by contempt for the people doing the work, most of whom are conscientious and constrained. It is served by describing the situation plainly and then asking what can actually be moved.